The Fasted 50K and Heavy Cream Cholesterol Experiment - Preview

 

The purpose of this piece is to preview a bit of an experiment I’m intending to perform in the near future. I’m calling it the “50k and Heavy Cream Cholesterol Experiment” because, well…I’m going to run 50 kilometers and drink a lot of heavy cream and sample lipid levels a number of times. Read on for details, reasoning, and predictions.

The Plan

The plan, broadly speaking, is to asses the effects of both an excessive dose of running and excessive saturated fat consumption on my lipid levels (LDL-C, HDL-C, triglycerides). These two interventions won’t be concurrent, but stacked immediately on top of one another over the course of a handful of days. Ideally, the plan is to begin this next Monday, January 20^th. I say ideally because a major factor in the timing is finding a day when I’m healthy enough to even run the 50k. As I’ve described recently, I still commonly miss days of exercise (and work) due to neurological complications. And of the days I’m healthy enough to get out the door to run, on exceedingly few could I reasonably hope to run a reliably strong 50k. This makes finding a good opportunity to carry out this experiment in the course of normal day to day life difficult at best. But as luck would have it, I’m on vacation this week and have a reasonable expectation of feeling pretty good when I return.

So, fly home on Sunday the 19^th and run the 50k on the 20^th. Beginning that evening and continuing for 3 full days after, I will consume a purely carnivore diet with as much saturated fat and dietary cholesterol as I can tolerate. The aim will be to consume several thousand calories per day above baseline, but exact numbers will depend on how exactly it feels to so greatly overindulge multiple days in a row (Despite the name I used in the title, I will not be consuming only heavy cream. Massive quantities of it, yes, but also meat, cheese, and butter). Baseline diet, for the record, is an animal-based ketogenic diet averaging about 80% calories from fat and fewer than 10g of carbohydrate per day.

The planned schedule is as follows:

Monday AM: Lipid Panel #1/Baseline

Monday AM: 50 kilometer run

Monday PM: Lipid Panel #2

Monday PM – Thursday PM: Heavy saturated fat consumption

Tuesday AM: Lipid Panel #3

Tuesday PM: Lipid Panel #4 (non-fasted)

Wednesday AM: Lipid Panel #5

Thursday AM: Lipid Panel #6

Friday AM: Lipid Panel #7/Final

 

What I’m Hoping to Measure

As you almost certainly know, the traditional paradigms surrounding diet and cholesterol suggest that consuming too much saturated fat and dietary cholesterol drives an increase in serum LDL cholesterol levels (in turn considered to be the prime driver of atherosclerotic cardiovascular disease). I, however, object to that paradigm, believing instead that the greatest factor influencing LDL-C levels is the body’s reliance on lipoproteins as an important delivery system.

Probably the most important cargo that lipoproteins carry are triglycerides, to be either stored as body fat or used as an energy source by the body. Which brings me to an important caveat that I’ve yet to mention – the 50 kilometer run will be carried out entirely in the fasted state. I will be consuming exactly zero calories before or during the run, not eating anything on the day until after my post-run blood draw.

This is a fairly extreme measure of course. Exceedingly few people ever run that far in a fasted state, and ever fewer (possibly zero?) have ever measured the effect of that effort on lipid levels. The American Heart Association and others suggest that saturated fat consumption is the greatest factor in raising cholesterol levels, with a lack of exercise a strong contender for number two. Conventual wisdom also tends to suggest that LDL-C levels don’t change rapidly, but instead over weeks or even months. It would stand to reason, then, that LDL-C should probably be largely unchanged between my first and second blood draws. Perhaps they might even tick down a fraction, as the intervening hours between the first and second blood draws will maximize typical guidelines for lowering cholesterol (plenty of exercise, zero fat consumption). If instead LDL-C increased during the run, it might require an update, or at least a caveat attached, to the typical paradigm.

Lets skip now to the final blood draw. This is, clearly, the extreme opposite end of the spectrum with respect to traditional cholesterol risk factors. I won’t exercise the three days between the 50k and the final blood draw, but I will eat so, so much saturated fat. And its flipping so aggressively from one extreme to the other that makes this fun. Again, a traditional medical mindset would suggest that LDL-C should clearly increase throughout the week as I binge saturated fat and dietary cholesterol. It may not increase a lot, as its only for a few days, but one would certainly expect it to start trending up in the face of such prodigious fat consumption (Just for fun – the AHA recommends capping saturated fat intake at ~13 grams per day. I intend to consume 25-30 times more than that each day. Essentially a month’s “worth” of saturated fat per day). So again, if the so-called expected outcome is not observed, it may suggest a shortcoming of the current conventional wisdom.

I’ll further expand on the day 2 blood draws momentarily, but the intervening lipid panels are largely to track trends throughout the week. I intended to skip the middle three blood draws at first, as its really the first and last days that will capture the full effect, but decided it would be more interesting to have a more complete dataset.

 

Predictions

Baseline/LP1 – I will have, by conventional standards, elevated LDL-C at baseline. I don’t know how elevated necessarily, but certainly it will be a number that would concern your average physician. On the contrary, I expect reasonably high HDL-C and low triglycerides that would be quite good by conventional standards. All of these values derive from the fact that I am a metabolically healthy individual consuming an exceedingly low-carbohydrate diet and thus relying on fatty acids for energy.  

LP2 – I expect LDL-C to rise fairly noticeably during the course of the fasted 50 kilometer run. Reliance on stored body fat for energy (or really, the hormonal effects of fasted exercise) will drive a significant increase in the breakdown of stored body fat, which should be largely trafficked through the liver and packaged in VLDL particles. The triglycerides in these VLDL particles will be taken up extremely rapidly by working muscles, causing the VLDL to convert to longer-lived LDL particles. This continuous effect will cause there to be an acute increase in cholesterol containing LDL particles, and thus an increase in measured LDL-C. In addition, I expect measured triglycerides to be extremely low for the same reason (most likely below my “personal best” of 66 mg/dl) as my working muscles rapidly take them up for energy.

Final/LP7 – The expectation here is that this result will also defy conventional wisdom. Not only will the extreme consumption of fatty animal products fail to raise my LDL-C, it will acutely lower levels to below baseline. Rather than relying heavily on stored body fat for energy, I’ll be doing the exact opposite. I’ll be creating a hormonal environment that more heavily emphasizes the storage of fat rather than its breakdown, thereby reducing the production of VLDL particles that would typically move my stored triglycerides around my body. Fewer VLDL particles means fewer LDL particles and thus lower LDL-C. Its worth noting, however, that this effect won’t be as great as it could be due to the compressed timeframe of this experiment. The average lifespan on an LDL particle is in the three and a half day range, and three and a half days before my final blood draw I’ll be producing huge number of VLDL/LDL particles during and immediately after my fasted run. A couple more days of binging would ensure these excess particles would be completely recycled, but frankly I don’t want to do this for that long, so…

Day 2/LP3 – Saving the best for last. This is, to me, the real meat of my experiment. I have strong preconceived assumptions about how the fasted exercise and the fat binge will effect lipids, but the blood draw on Tuesday morning is for me the one that ventures into the great unknown. And frankly, in a lot of ways, it ventures into the collective scientific unknown, as I don’t think anybody has ever documented the effects of such an extreme scenario on lipid levels.

Let’s first asses what this blood draw might look like if we only consider the energy deliver nature of lipids. Remember again that LDL particles have a typical lifespan of 3+ days. This blood draw, maybe 17 hours after the second, will represent only ~20 percent of the lifespan of a typical LDL particle. And while the massive effort between the first two blood draws should generate a significant acute increase in LDL particles, nothing about the rest and recovery after lipid panel 2 should differ greatly from what I’d be doing three to four days earlier. That is to say, there shouldn’t be much reason for the number of particles produced to differ greatly from the number being recycled. It may even be the case that energy demand remains so high in the immediate aftermath of the run and the second blood draw that LDL-C could fractionally increase if I don’t eat enough or quickly enough to fully blunt that effect. So, from a purely energy driven perspective, LDL-C levels at or just above those in lipid panel 2 might be reasonably expected (with triglycerides returning closer to baseline as well).

But…what if energy (and cholesterol) weren’t the only important components being trafficked by lipoproteins? What if another effect were present that could also drive a noticeable change in LDL-C levels? This, essentially, is what I’m hoping to test.

It may be that a very important and underappreciated element that LDL particles transport…is just themselves. After all, lipoproteins are made largely of the same phospholipids that comprise cell membranes throughout the human body. And it could very well be the case that an acute insult to enough of those cell membranes – for example, the damage caused by running 50 kilometers – could cause many LDL particles to be taken up by the cells as raw materials for the repair of these damaged membranes (and/or the creation of new ones).

If this were the case, a reasonable proportion of the existing LDL particles in circulation might leave the bloodstream earlier than expected, thus decreasing LDL-C from the energy driven expectation outlined just above.

To be clear, I don’t have a reasonable guess for what my LDL-C will look like on Tuesday morning. Something wildly different than expected on the post-run or post-binge panels would require some reevaluation of the energy delivery paradigm. However, I’m not making any particular prediction for this lipid panel. I do strongly believe, however, that a decrease in LDL-C from lipid panel 2 to panel 3 would be indicative only of this proposed effect – the endocytosis of LDL particles for the repair of cellular damage. And I think demonstrating this effect would, in theory, go a long ways towards further understanding a transport model of lipoprotein function and even the underlying causes of atherosclerotic cardiovascular disease. If that decrease is in fact observed, I’ll of course have plenty to say about it after the fact.

Summary

So, there you have, in two thousand words – a weeklong experiment to test the extremes of lipid mechanics and assess the ways in which a lipid transport system may best explain lipid behavior. To the best of my knowledge, this is a novel demonstration, at least at this extreme. Studies have demonstrated that a great energy deficit raises LDL-C, and numerous individuals (myself included) have lowered LDL-C while binging on fat. But the extreme, hyper-condensed nature of this N=1 experiment is, I think, without parallel. In particular, the second day’s blood draw, on the back of a such a significant physiological event, has the potential to demonstrate a possible underappreciated characteristic of lipid behavior in the human body. Whether this ultimately demonstrates something significantly novel, or only highlights the importance of lipids in energy deliver, or goes up in flames entirely, remains to be seen. But, regardless, results and summaries should come soon after. To be continued.

Summer Health Update Part 2 - What 3 Pounds of Meat Every Day Does To Your Blood Work

 

While my neurological health remains a few percentage points short of ideal, I’ve also been increasingly in touch with other aspects of my health as I continue to seek optimal recovery. The amount of “general health” bloodwork I’ve sought, measured, and ordered has increased significantly and, with that in mind, I’ve decided to share the most recent round of (semi-) comprehensive of blood work.

Conventional wisdom suggests that eating a diet high in meat and fat is dangerous for cardiometabolic and chronic health. For over a decade, I’ve progressively ignored that conventional wisdom. For years, this meant eating a lower carb (for an athlete, at least) paleo-type diet. Since my vaccine reaction, it has meant consuming various degrees of low-carb and ketogenic diets. And for more than a year now, it has meant more than 90% of my diet as beef. While I was largely sedentary for the first two years of my illness, the last several months have featured increasing amounts of exercise.

Despite the extended lack of activity and the ostensibly hazardous dietary reliance on meat and fat, you can see below that my chronic health markers are generally quite good. Below are the relevant results from my latest check-in and my commentary on each of the metrics. My recent diet and exercise statistics for reference:

Background statistics - 

Height – 6ft 3in        Weight – 161 lbs.         Age - 32y, 7m

     

8 week exercise averages –

~47 miles/week hiking and jogging, 2-3/week ~20 min strength training

8 week dietary averages –

~3320 calories/day        

70.5% fat/27.1% protein/2.4% carbohydrate (~19g/day)

48% saturated fat/48% monounsaturated fat/4% polyunsaturated fat

3 day dietary averages –

~3430 calories/day

71.2% fat/23.7% protein/5.1% carbohydrate (~43g/day)

48% sfa/48% mufa/4% pufa

 

 

 

 

Triglycerides – 87 mg/dL (Reference range 0-149)

I eat tons of fat, but don’t have tons of fat moving around my blood. What gives? Well, your triglyceride levels don’t reflect fat consumption or triglyceride production by the liver. Trigs reflect fatty acid utilization and fat metabolism. If you efficiently metabolize/utilize fat for energy, you should have low triglycerides. Common recommendations suggest under 150 to be normal, but realistically 150 is pretty sketchy and any values north of 100 suggest room for improvement.

 

HDL – 68 mg/dL (Reference range >39)

HDL levels are primarily responsive to two factors – triglyceride levels, and fat consumption. Elevated triglycerides resulting from metabolic inefficiency subsequently lead to a reduction in HDL-C (you can read about why here!). Meanwhile, fat consumption directly increases the concentration of the structural lipoproteins that eventually form HDL particles. Ergo, low trigs + high fat consumption = high HDL.

 

Triglyceride/HDL Ratio – 1.28

Not a unique measurement, but a reasonably meaningful reflection of metabolic health. Because poor metabolic health increases triglycerides and subsequently decreases HDL, a ratio between the two is a decent proxy for metabolic health. Conventional wisdom would suggest something like 3.5 to still be a fine value, even though cardiometabolic disease rates start exploding once you inch above this level. Personally, I wouldn’t feel great about anything higher than 1.5-2. For fun, I’ll include comparisons to bloodwork taken the day after a brief spring “binge” (3 days of higher carb, higher calorie consumption) as well as the day I ended an 8 day extended fast.

	April “Binge”	April Fast	June
Triglycerides	127	84	87
HDL-C	50	52	68
Trig/HDL Ratio	2.54	1.61	1.28

 

LDL – 113 mg/dL (Reference range 0-99)

Just about the least meaningful standalone marker out there, despite the medical and pharmaceutical indu$try’$ endle$$ obe$$e$$ion with $elling $tatins in order to force it lower in basically everyone. You can read tens of thousands of words I’ve written about the problems with LDL here or here if you’re interested. I’ll say this for now though – LDL is hyper-agile in a metabolically healthy person. My values can effortlessly bounce between roughly 100 and 200 depending on what I eat on any given day. I don’t care to ever see numbers lower than that, as I see no benefit whatsoever (and, for whatever its worth, low LDL is associated with significant increase in death and disease for several plausible reasons). Furthermore, you’ll note that despite consuming tons of fat and saturated fat, this value is actually slightly below the population average, even if its slightly above the recommended level. That’s because saturated fat is absolutely not the prime driver of LDL levels.

 

Apolipoprotein B – 84 mg/dL (Reference range <90)

ApoB is the structural protein that forms LDL particles, and is slowly beginning to replace LDL as the en vouge cardiovascular risk measure (It is a better measure than LDL, but is subject to many of the same flaws as well). This value reflects the number of LDL particles in circulation, and you’ll note once again that despite eating tons of fat my values are actually below average and in the “approved” medical range. 

 

LDL/ApoB Ratio – 1.35

My ApoB, which reflects LDL particle count, is in the recommended range, but my LDL cholesterol is still high. How does this work? The answer lies in particle size – fewer ApoB particles carrying a given amount of cholesterol suggests those particles are on the larger size. This matters for a couple reasons – small particles indicate poor metabolic efficiency, while being themselves highly susceptible to the oxidative and glycemic damage that commonly triggers the immune-mediated atherosclerotic process. A ratio of 1.2 or so is a common cut point in the literature, with ratios below that suggesting significant cardiometabolic risk. I’ve previously forced mine as low as 1.15 with just a couple days of higher carb consumption, but would prefer not to see values below about 1.3 in typical conditions. Prior comparisons included here as well.

	April “Binge”	April Fast	June
LDL-C	108	180	113
ApoB	94	139	84
LDL/ApoB Ratio	1.15	1.29	1.35

 

C-Reactive Protein - <1 mg/L (Reference range 0-10)

CRP is a measure of systemic inflammation. You’d like to see this number as close to zero as possible, generally speaking, and the reference range extending to 10 is flat-out crazy. In the absence of some other relevant factor like a recent race, I’d really hate to see even a value of 2. Unfortunately, LabCorp doesn’t report values below 1, meaning you never really want to see an actual value on one of these tests. The one time I managed to have this tested at another lab, it was at 0.2.

 

Hemoglobin A1C – 5.1% (Reference range 4.8 - 5.6)

HbA1C is a measure of long-term blood sugar (specifically a measure of how many red blood cells have been glycated by sugar in the blood). Its commonly used to assess or monitor diabetes status. Values for HbA1C exist across a fairly narrow band – 5 is great, 6 is pretty terrible (though plenty of people hit 8, 9, or even higher). Current guidelines consider 5.7 or higher to be “prediabetes” and you really don’t want to see this above the low 5s.

 

Glucose – 95 mg/dL (Reference range 0-99)

This is on the high side for me, as fasting glucose usually bounces around between about 85 and 95. I don’t think a single number is worth all that much when you can just look at A1C and capture a long-term picture, but it’s a normal enough number regardless

 

Insulin – 1.7 uIU/mL (Reference range 2.6-24.9)

Arguably the single most important measurement on here in my view. Insulin is a storage and growth hormone secreted primarily in response to carbohydrate consumption. Chronically elevated levels of insulin are instrumental in metabolic dysfunction and contribute to the insulin resistance that defines diabetes and so much of cardiometabolic disease. The normal reference range of “less that 25” is absolutely off the rails. A person with fasting insulin levels in the 20s is so metabolically sick. Just ridiculous to label it normal in any sense of the word. This is a number you want in the low to mid single digits, with numbers closer to 10 more than sufficient to disrupt optimal metabolic health and function. As mentioned, carbohydrates are the primary driver of insulin levels. I consume very few, and thus have very low fasting insulin.

 

HOMA-IR - 0.4

The Homeostatic Model Assessment of Insulin Resistance is a simple, non-invasive method of estimating an individual’s resistance to insulin using fasting glucose and insulin values. Insulin resistance is a prime driver of heart disease and other chronic diseases, and quite literally is diabetes. HOMA-IR values under 1 are considered optimal, with values north of 2 indicating moderate or greater insulin resistance. A low HOMA-IR and high insulin sensitivity are generally to be expected when consuming a low-carbohydrate diet. I'll add the binge/fast comparison here as well

	April “Binge”	April Fast	June
Glucose	106	66	95
Insulin	11.4	1.3	1.7
HOMA-IR	3.0	0.2	0.4

 

Uric Acid – 3.5 mg/dL (Reference range 3.8-8.4)

Say it with me – “red meat doesn’t cause gout.” This is bit of nonsense that continues to be propagated throughout nutrition and medical circles, but it doesn’t reflect reality. Uric acid is a nitrogen-containing compound that forms from the breakdown of purines, which are indeed found more abundantly in animal products than in plants. But then a healthy person just pees the uric acid out, while a metabolically dysfunctional individual will not. Which is why elevated uric acid levels are tightly linked to insulin levels, obesity, and metabolic syndrome, while mine is out the bottom of LabCorp’s reference range.

 

Vitamin D – 39.1 ng/mL (Reference range 30-100)

This is lower than I’d like. The reference range says above 30 is fine but would realistically like to be double that. I already triggered neuro symptoms trying some vitamin D drops so now the strategy will be a bit more eggs, salmon, and mid-day sun before maybe assessing again.

 

Thyroxine (T4) – 1.2 ng/dL, TSH – 1.14 uIU/mL (Reference range 0.82-1.77, 0.45-4.5) 

My thyroid hormones are perfectly normal.

 

Blood pressure – 110/70, 110/64 mmHg (Reference range <120/80)

These are my two latest doctor’s office BP readings, although I somewhat regularly measure my own BP and find these values to be quite typical. Elevated blood pressure is really just another manifestation of chronic insulin resistance, rather than salt consumption or any other acute dietary factor (I literally drink salt in my water for whatever that’s worth). Its only chronic carb/sugar consumption and elevated insulin that will raise blood pressure, so again optimal measures are unsurprising.

 

Testosterone: Total – 183 ng/dL, Free - 4.4 pg/mL (Reference range 264-916, 8.7-25.1)

And here’s the one that was actually a problem. Normal testosterone for a healthy 30-something should be a few hundred points higher than this. This proved to be a big sign that I wasn’t eating enough, as downregulated hormone production is one obvious consequence of underfueling (this is just one reason that “calories” is a quasi-worthless way to approach weight and metabolism). Unsurprisingly, I was more sluggish than I should have been and slowly losing weight as well. But while my testosterone was quite low, the good news is I’ve since rectified the problem pretty much by just eating more. A month later, my total testosterone was up to 543.

There you have yet...meat and fat aren't killing me yet!